1 | /*************************************************************************
|
---|
2 | * This file & class is part of the StarMath Project
|
---|
3 | * Copyright 2010, 2011 Matthew Ira Campbell, PhD.
|
---|
4 | *
|
---|
5 | * StarMath is free software: you can redistribute it and/or modify
|
---|
6 | * it under the terms of the GNU General Public License as published by
|
---|
7 | * the Free Software Foundation, either version 3 of the License, or
|
---|
8 | * (at your option) any later version.
|
---|
9 | *
|
---|
10 | * StarMath is distributed in the hope that it will be useful,
|
---|
11 | * but WITHOUT ANY WARRANTY; without even the implied warranty of
|
---|
12 | * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
---|
13 | * GNU General Public License for more details.
|
---|
14 | *
|
---|
15 | * You should have received a copy of the GNU General Public License
|
---|
16 | * along with StarMath. If not, see <http://www.gnu.org/licenses/>.
|
---|
17 | *
|
---|
18 | * Please find further details and contact information on StarMath
|
---|
19 | * at http://starmath.codeplex.com/.
|
---|
20 | *************************************************************************/
|
---|
21 | using System;
|
---|
22 | using System.Collections.Generic;
|
---|
23 | using System.Linq;
|
---|
24 |
|
---|
25 |
|
---|
26 | namespace MIConvexHull
|
---|
27 | {
|
---|
28 | public static class StarMath
|
---|
29 | {
|
---|
30 | #region Printing
|
---|
31 | private const int cellWidth = 10;
|
---|
32 | private const int numDecimals = 3;
|
---|
33 |
|
---|
34 | /// <summary>
|
---|
35 | /// Makes the print string.
|
---|
36 | /// </summary>
|
---|
37 | /// <param name = "A">The A.</param>
|
---|
38 | public static string MakePrintString(double[,] A)
|
---|
39 | {
|
---|
40 | if (A == null) return "<null>";
|
---|
41 | var format = "{0:F" + numDecimals + "}";
|
---|
42 | var p = "";
|
---|
43 | for (var i = 0; i < A.GetLength(0); i++)
|
---|
44 | {
|
---|
45 | p += "| ";
|
---|
46 | for (var j = 0; j < A.GetLength(1); j++)
|
---|
47 | p += formatCell(format, A[i, j]) + ",";
|
---|
48 | p = p.Remove(p.Length - 1);
|
---|
49 | p += " |\n";
|
---|
50 | }
|
---|
51 | p = p.Remove(p.Length - 1);
|
---|
52 | return p;
|
---|
53 | }
|
---|
54 |
|
---|
55 | private static object formatCell(string format, double p)
|
---|
56 | {
|
---|
57 | var numStr = string.Format(format, p);
|
---|
58 | numStr = numStr.TrimEnd('0');
|
---|
59 | numStr = numStr.TrimEnd('.');
|
---|
60 | var padAmt = ((double)(cellWidth - numStr.Length)) / 2;
|
---|
61 | numStr = numStr.PadLeft((int)Math.Floor(padAmt + numStr.Length));
|
---|
62 | numStr = numStr.PadRight(cellWidth);
|
---|
63 | return numStr;
|
---|
64 | }
|
---|
65 | #endregion
|
---|
66 |
|
---|
67 | #region Solve
|
---|
68 | /// <summary>
|
---|
69 | /// Solves the specified A matrix.
|
---|
70 | /// </summary>
|
---|
71 | /// <param name="A">The A.</param>
|
---|
72 | /// <param name="b">The b.</param>
|
---|
73 | /// <returns></returns>
|
---|
74 | public static double[] solve(double[,] A, IList<double> b)
|
---|
75 | {
|
---|
76 | var aLength = A.GetLength(0);
|
---|
77 | if (aLength != A.GetLength(1))
|
---|
78 | throw new Exception("Matrix, A, must be square.");
|
---|
79 | if (aLength != b.Count)
|
---|
80 | throw new Exception("Matrix, A, must be have the same number of rows as the vector, b.");
|
---|
81 |
|
---|
82 | /****** need code to determine when to switch between *****
|
---|
83 | ****** this analytical approach and the SOR approach *****/
|
---|
84 | return multiply(inverse(A, aLength), b, aLength, aLength);
|
---|
85 | }
|
---|
86 |
|
---|
87 | public static void solveDestructiveInto(double[,] A, double[] b, double[] target)
|
---|
88 | {
|
---|
89 | var aLength = A.GetLength(0);
|
---|
90 | if (aLength != A.GetLength(1))
|
---|
91 | throw new Exception("Matrix, A, must be square.");
|
---|
92 | if (aLength != b.Length)
|
---|
93 | throw new Exception("Matrix, A, must be have the same number of rows as the vector, b.");
|
---|
94 |
|
---|
95 | /****** need code to determine when to switch between *****
|
---|
96 | ****** this analytical approach and the SOR approach *****/
|
---|
97 | inverseInPlace(A, aLength);
|
---|
98 | multiplyInto(A, b, aLength, aLength, target);
|
---|
99 | }
|
---|
100 |
|
---|
101 | /// <summary>
|
---|
102 | /// does gaussian elimination.
|
---|
103 | /// </summary>
|
---|
104 | /// <param name="nDim"></param>
|
---|
105 | /// <param name="pfMatr"></param>
|
---|
106 | /// <param name="pfVect"></param>
|
---|
107 | /// <param name="pfSolution"></param>
|
---|
108 | public static void gaussElimination(int nDim, double[][] pfMatr, double[] pfVect, double[] pfSolution)
|
---|
109 | {
|
---|
110 | double fMaxElem;
|
---|
111 | double fAcc;
|
---|
112 |
|
---|
113 | int i, j, k, m;
|
---|
114 |
|
---|
115 | for (k = 0; k < (nDim - 1); k++) // base row of matrix
|
---|
116 | {
|
---|
117 | var rowK = pfMatr[k];
|
---|
118 |
|
---|
119 | // search of line with max element
|
---|
120 | fMaxElem = Math.Abs(rowK[k]);
|
---|
121 | m = k;
|
---|
122 | for (i = k + 1; i < nDim; i++)
|
---|
123 | {
|
---|
124 | if (fMaxElem < Math.Abs(pfMatr[i][k]))
|
---|
125 | {
|
---|
126 | fMaxElem = pfMatr[i][k];
|
---|
127 | m = i;
|
---|
128 | }
|
---|
129 | }
|
---|
130 |
|
---|
131 | // permutation of base line (index k) and max element line(index m)
|
---|
132 | if (m != k)
|
---|
133 | {
|
---|
134 | var rowM = pfMatr[m];
|
---|
135 | for (i = k; i < nDim; i++)
|
---|
136 | {
|
---|
137 | fAcc = rowK[i];
|
---|
138 | rowK[i] = rowM[i];
|
---|
139 | rowM[i] = fAcc;
|
---|
140 | }
|
---|
141 | fAcc = pfVect[k];
|
---|
142 | pfVect[k] = pfVect[m];
|
---|
143 | pfVect[m] = fAcc;
|
---|
144 | }
|
---|
145 |
|
---|
146 | //if( pfMatr[k*nDim + k] == 0.0) return 1; // needs improvement !!!
|
---|
147 |
|
---|
148 | // triangulation of matrix with coefficients
|
---|
149 | for (j = (k + 1); j < nDim; j++) // current row of matrix
|
---|
150 | {
|
---|
151 | var rowJ = pfMatr[j];
|
---|
152 | fAcc = -rowJ[k] / rowK[k];
|
---|
153 | for (i = k; i < nDim; i++)
|
---|
154 | {
|
---|
155 | rowJ[i] = rowJ[i] + fAcc * rowK[i];
|
---|
156 | }
|
---|
157 | pfVect[j] = pfVect[j] + fAcc * pfVect[k]; // free member recalculation
|
---|
158 | }
|
---|
159 | }
|
---|
160 |
|
---|
161 | for (k = (nDim - 1); k >= 0; k--)
|
---|
162 | {
|
---|
163 | var rowK = pfMatr[k];
|
---|
164 | pfSolution[k] = pfVect[k];
|
---|
165 | for (i = (k + 1); i < nDim; i++)
|
---|
166 | {
|
---|
167 | pfSolution[k] -= (rowK[i] * pfSolution[i]);
|
---|
168 | }
|
---|
169 | pfSolution[k] = pfSolution[k] / rowK[k];
|
---|
170 | }
|
---|
171 | }
|
---|
172 |
|
---|
173 | #endregion
|
---|
174 |
|
---|
175 | #region inverse transpose
|
---|
176 |
|
---|
177 | /// <summary>
|
---|
178 | /// Inverses the matrix A only if the diagonal is all non-zero.
|
---|
179 | /// A[i,i] != 0.0
|
---|
180 | /// </summary>
|
---|
181 | /// <param name = "A">The matrix to invert. This matrix is unchanged by this function.</param>
|
---|
182 | /// <param name="length">The length of the number of rows/columns in the square matrix, A.</param>
|
---|
183 | /// <returns>The inverted matrix, A^-1.</returns>
|
---|
184 | public static double[,] inverse(double[,] A, int length)
|
---|
185 | {
|
---|
186 | if (length == 1) return new[,] { { 1 / A[0, 0] } };
|
---|
187 | return inverseWithLUResult(LUDecomposition(A, length), length);
|
---|
188 | }
|
---|
189 |
|
---|
190 | public static void inverseInPlace(double[,] A, int length)
|
---|
191 | {
|
---|
192 | if (length == 1)
|
---|
193 | {
|
---|
194 | A[0, 0] = 1 / A[0, 0];
|
---|
195 | return;
|
---|
196 | }
|
---|
197 | LUDecompositionInPlace(A, length);
|
---|
198 | inverseWithLUResult(A, length);
|
---|
199 | }
|
---|
200 |
|
---|
201 |
|
---|
202 | /// <summary>
|
---|
203 | /// Returns the LU decomposition of A in a new matrix.
|
---|
204 | /// </summary>
|
---|
205 | /// <param name = "A">The matrix to invert. This matrix is unchanged by this function.</param>
|
---|
206 | /// <param name="length">The length of the number of rows/columns in the square matrix, A.</param>
|
---|
207 | /// <returns>A matrix of equal size to A that combines the L and U. Here the diagonals belongs to L and the U's diagonal elements are all 1.</returns>
|
---|
208 | public static double[,] LUDecomposition(double[,] A, int length)
|
---|
209 | {
|
---|
210 | var B = (double[,])A.Clone();
|
---|
211 | // normalize row 0
|
---|
212 | for (var i = 1; i < length; i++) B[0, i] /= B[0, 0];
|
---|
213 |
|
---|
214 | for (var i = 1; i < length; i++)
|
---|
215 | {
|
---|
216 | for (var j = i; j < length; j++)
|
---|
217 | {
|
---|
218 | // do a column of L
|
---|
219 | var sum = 0.0;
|
---|
220 | for (var k = 0; k < i; k++)
|
---|
221 | sum += B[j, k] * B[k, i];
|
---|
222 | B[j, i] -= sum;
|
---|
223 | }
|
---|
224 | if (i == length - 1) continue;
|
---|
225 | for (var j = i + 1; j < length; j++)
|
---|
226 | {
|
---|
227 | // do a row of U
|
---|
228 | var sum = 0.0;
|
---|
229 | for (var k = 0; k < i; k++)
|
---|
230 | sum += B[i, k] * B[k, j];
|
---|
231 | B[i, j] =
|
---|
232 | (B[i, j] - sum) / B[i, i];
|
---|
233 | }
|
---|
234 | }
|
---|
235 | return B;
|
---|
236 | }
|
---|
237 |
|
---|
238 | public static void LUDecompositionInPlace(double[,] A, int length)
|
---|
239 | {
|
---|
240 | var B = A;
|
---|
241 | // normalize row 0
|
---|
242 | for (var i = 1; i < length; i++) B[0, i] /= B[0, 0];
|
---|
243 |
|
---|
244 | for (var i = 1; i < length; i++)
|
---|
245 | {
|
---|
246 | for (var j = i; j < length; j++)
|
---|
247 | {
|
---|
248 | // do a column of L
|
---|
249 | var sum = 0.0;
|
---|
250 | for (var k = 0; k < i; k++)
|
---|
251 | sum += B[j, k] * B[k, i];
|
---|
252 | B[j, i] -= sum;
|
---|
253 | }
|
---|
254 | if (i == length - 1) continue;
|
---|
255 | for (var j = i + 1; j < length; j++)
|
---|
256 | {
|
---|
257 | // do a row of U
|
---|
258 | var sum = 0.0;
|
---|
259 | for (var k = 0; k < i; k++)
|
---|
260 | sum += B[i, k] * B[k, j];
|
---|
261 | B[i, j] =
|
---|
262 | (B[i, j] - sum) / B[i, i];
|
---|
263 | }
|
---|
264 | }
|
---|
265 | }
|
---|
266 |
|
---|
267 | private static double[,] inverseWithLUResult(double[,] B, int length)
|
---|
268 | {
|
---|
269 | // this code is adapted from http://users.erols.com/mdinolfo/matrix.htm
|
---|
270 | // one constraint/caveat in this function is that the diagonal elts. cannot
|
---|
271 | // be zero.
|
---|
272 | // if the matrix is not square or is less than B 2x2,
|
---|
273 | // then this function won't work
|
---|
274 | #region invert L
|
---|
275 |
|
---|
276 | for (var i = 0; i < length; i++)
|
---|
277 | for (var j = i; j < length; j++)
|
---|
278 | {
|
---|
279 | var x = 1.0;
|
---|
280 | if (i != j)
|
---|
281 | {
|
---|
282 | x = 0.0;
|
---|
283 | for (var k = i; k < j; k++)
|
---|
284 | x -= B[j, k] * B[k, i];
|
---|
285 | }
|
---|
286 | B[j, i] = x / B[j, j];
|
---|
287 | }
|
---|
288 |
|
---|
289 | #endregion
|
---|
290 |
|
---|
291 | #region invert U
|
---|
292 |
|
---|
293 | for (var i = 0; i < length; i++)
|
---|
294 | for (var j = i; j < length; j++)
|
---|
295 | {
|
---|
296 | if (i == j) continue;
|
---|
297 | var sum = 0.0;
|
---|
298 | for (var k = i; k < j; k++)
|
---|
299 | sum += B[k, j] * ((i == k) ? 1.0 : B[i, k]);
|
---|
300 | B[i, j] = -sum;
|
---|
301 | }
|
---|
302 |
|
---|
303 | #endregion
|
---|
304 |
|
---|
305 | #region final inversion
|
---|
306 |
|
---|
307 | for (var i = 0; i < length; i++)
|
---|
308 | for (var j = 0; j < length; j++)
|
---|
309 | {
|
---|
310 | var sum = 0.0;
|
---|
311 | for (var k = ((i > j) ? i : j); k < length; k++)
|
---|
312 | sum += ((j == k) ? 1.0 : B[j, k]) * B[k, i];
|
---|
313 | B[j, i] = sum;
|
---|
314 | }
|
---|
315 |
|
---|
316 | #endregion
|
---|
317 |
|
---|
318 | return B;
|
---|
319 | }
|
---|
320 | /// <summary>
|
---|
321 | /// Returns the determinant of matrix, A.
|
---|
322 | /// </summary>
|
---|
323 | /// <param name = "A">The input argument matrix. This matrix is unchanged by this function.</param>
|
---|
324 | /// <exception cref = "Exception"></exception>
|
---|
325 | /// <returns>a single value representing the matrix's determinant.</returns>
|
---|
326 | public static double determinant(double[,] A)
|
---|
327 | {
|
---|
328 | if (A == null) throw new Exception("The matrix, A, is null.");
|
---|
329 | var length = A.GetLength(0);
|
---|
330 | if (length != A.GetLength(1))
|
---|
331 | throw new Exception("The determinant is only possible for square matrices.");
|
---|
332 | if (length == 0) return 0.0;
|
---|
333 | if (length == 1) return A[0, 0];
|
---|
334 | if (length == 2) return (A[0, 0] * A[1, 1]) - (A[0, 1] * A[1, 0]);
|
---|
335 | if (length == 3)
|
---|
336 | return (A[0, 0] * A[1, 1] * A[2, 2])
|
---|
337 | + (A[0, 1] * A[1, 2] * A[2, 0])
|
---|
338 | + (A[0, 2] * A[1, 0] * A[2, 1])
|
---|
339 | - (A[0, 0] * A[1, 2] * A[2, 1])
|
---|
340 | - (A[0, 1] * A[1, 0] * A[2, 2])
|
---|
341 | - (A[0, 2] * A[1, 1] * A[2, 0]);
|
---|
342 | return determinantBig(A, length);
|
---|
343 | }
|
---|
344 |
|
---|
345 | /// <summary>
|
---|
346 | /// Modifies the matrix during the computation if the dimension > 3.
|
---|
347 | /// </summary>
|
---|
348 | /// <param name="A"></param>
|
---|
349 | /// <returns></returns>
|
---|
350 | public static double determinantDestructive(double[,] A, int dimension)
|
---|
351 | {
|
---|
352 | if (A == null) throw new Exception("The matrix, A, is null.");
|
---|
353 | var length = dimension;
|
---|
354 | //if (length != A.GetLength(1))
|
---|
355 | // throw new Exception("The determinant is only possible for square matrices.");
|
---|
356 | if (length == 0) return 0.0;
|
---|
357 | if (length == 1) return A[0, 0];
|
---|
358 | if (length == 2) return (A[0, 0] * A[1, 1]) - (A[0, 1] * A[1, 0]);
|
---|
359 | if (length == 3)
|
---|
360 | return (A[0, 0] * A[1, 1] * A[2, 2])
|
---|
361 | + (A[0, 1] * A[1, 2] * A[2, 0])
|
---|
362 | + (A[0, 2] * A[1, 0] * A[2, 1])
|
---|
363 | - (A[0, 0] * A[1, 2] * A[2, 1])
|
---|
364 | - (A[0, 1] * A[1, 0] * A[2, 2])
|
---|
365 | - (A[0, 2] * A[1, 1] * A[2, 0]);
|
---|
366 | return determinantBigDestructive(A, length);
|
---|
367 | }
|
---|
368 |
|
---|
369 | static double determinantBigDestructive(double[,] A, int length)
|
---|
370 | {
|
---|
371 | LUDecompositionInPlace(A, length);
|
---|
372 | var result = 1.0;
|
---|
373 | for (var i = 0; i < length; i++)
|
---|
374 | if (double.IsNaN(A[i, i]))
|
---|
375 | return 0;
|
---|
376 | else result *= A[i, i];
|
---|
377 | return result;
|
---|
378 | }
|
---|
379 |
|
---|
380 | /// <summary>
|
---|
381 | /// Returns the determinant of matrix, A. Only used internally for matrices larger than 3.
|
---|
382 | /// </summary>
|
---|
383 | /// <param name="A">The input argument matrix. This matrix is unchanged by this function.</param>
|
---|
384 | /// <param name="length">The length of the side of the square matrix.</param>
|
---|
385 | /// <returns>
|
---|
386 | /// a single value representing the matrix's determinant.
|
---|
387 | /// </returns>
|
---|
388 | public static double determinantBig(double[,] A, int length)
|
---|
389 | {
|
---|
390 | double[,] L, U;
|
---|
391 | LUDecomposition(A, out L, out U, length);
|
---|
392 | var result = 1.0;
|
---|
393 | for (var i = 0; i < length; i++)
|
---|
394 | if (double.IsNaN(L[i, i]))
|
---|
395 | return 0;
|
---|
396 | else result *= L[i, i];
|
---|
397 | return result;
|
---|
398 | }
|
---|
399 | /// <summary>
|
---|
400 | /// Returns the LU decomposition of A in a new matrix.
|
---|
401 | /// </summary>
|
---|
402 | /// <param name = "A">The matrix to invert. This matrix is unchanged by this function.</param>
|
---|
403 | /// <param name = "L">The L matrix is output where the diagonal elements are included and not (necessarily) equal to one.</param>
|
---|
404 | /// <param name = "U">The U matrix is output where the diagonal elements are all equal to one.</param>
|
---|
405 | /// <param name="length">The length of the number of rows/columns in the square matrix, A.</param>
|
---|
406 | public static void LUDecomposition(double[,] A, out double[,] L, out double[,] U, int length)
|
---|
407 | {
|
---|
408 | L = LUDecomposition(A, length);
|
---|
409 | U = new double[length, length];
|
---|
410 | for (var i = 0; i < length; i++)
|
---|
411 | {
|
---|
412 | U[i, i] = 1.0;
|
---|
413 | for (var j = i + 1; j < length; j++)
|
---|
414 | {
|
---|
415 | U[i, j] = L[i, j];
|
---|
416 | L[i, j] = 0.0;
|
---|
417 | }
|
---|
418 | }
|
---|
419 | }
|
---|
420 |
|
---|
421 | #endregion
|
---|
422 |
|
---|
423 | #region norm
|
---|
424 | /// <summary>
|
---|
425 | /// Returns to 2-norm (square root of the sum of squares of all terms)
|
---|
426 | /// of the vector, x.
|
---|
427 | /// </summary>
|
---|
428 | /// <param name="x">The vector, x.</param>
|
---|
429 | /// <param name="dontDoSqrt">if set to <c>true</c> [don't take the square root].</param>
|
---|
430 | /// <returns>Scalar value of 2-norm.</returns>
|
---|
431 | public static double norm2(double[] x, int dim, Boolean dontDoSqrt = false)
|
---|
432 | {
|
---|
433 | double norm = 0;
|
---|
434 | for (int i = 0; i < dim; i++)
|
---|
435 | {
|
---|
436 | var t = x[i];
|
---|
437 | norm += t * t;
|
---|
438 | }
|
---|
439 | return dontDoSqrt ? norm : Math.Sqrt(norm);
|
---|
440 | }
|
---|
441 | /// <summary>
|
---|
442 | /// Returns to normalized vector (has lenght or 2-norm of 1))
|
---|
443 | /// of the vector, x.
|
---|
444 | /// </summary>
|
---|
445 | /// <param name="x">The vector, x.</param>
|
---|
446 | /// <param name="length">The length of the vector.</param>
|
---|
447 | /// <returns>unit vector.</returns>
|
---|
448 | public static double[] normalize(double[] x, int length)
|
---|
449 | {
|
---|
450 | return divide(x, norm2(x, length), length);
|
---|
451 | }
|
---|
452 |
|
---|
453 | public static void normalizeInPlace(double[] x, int length)
|
---|
454 | {
|
---|
455 | double f = 1.0 / norm2(x, length);
|
---|
456 | for (int i = 0; i < length; i++) x[i] *= f;
|
---|
457 | }
|
---|
458 | #endregion
|
---|
459 |
|
---|
460 | #region make extract
|
---|
461 | /// <summary>
|
---|
462 | /// Gets a row of matrix, A.
|
---|
463 | /// </summary>
|
---|
464 | /// <param name = "rowIndex">The row index.</param>
|
---|
465 | /// <param name = "A">The matrix, A.</param>
|
---|
466 | /// <returns>A double array that contains the requested row</returns>
|
---|
467 | public static double[] GetRow(int rowIndex, double[,] A)
|
---|
468 | {
|
---|
469 | var numRows = A.GetLength(0);
|
---|
470 | var numCols = A.GetLength(1);
|
---|
471 | if ((rowIndex < 0) || (rowIndex >= numRows))
|
---|
472 | throw new Exception("MatrixMath Size Error: An index value of "
|
---|
473 | + rowIndex
|
---|
474 | + " for getRow is not in required range from 0 up to (but not including) "
|
---|
475 | + numRows + ".");
|
---|
476 | var v = new double[numCols];
|
---|
477 | for (var i = 0; i < numCols; i++)
|
---|
478 | v[i] = A[rowIndex, i];
|
---|
479 | return v;
|
---|
480 | }
|
---|
481 | /// <summary>
|
---|
482 | /// Sets/Replaces the given column of matrix A with the vector v.
|
---|
483 | /// </summary>
|
---|
484 | /// <param name = "colIndex">Index of the col.</param>
|
---|
485 | /// <param name = "A">The A.</param>
|
---|
486 | /// <param name = "v">The v.</param>
|
---|
487 | public static void SetColumn(int colIndex, double[,] A, IList<double> v)
|
---|
488 | {
|
---|
489 | var numRows = A.GetLength(0);
|
---|
490 | var numCols = A.GetLength(1);
|
---|
491 | if ((colIndex < 0) || (colIndex >= numCols))
|
---|
492 | throw new Exception("MatrixMath Size Error: An index value of "
|
---|
493 | + colIndex
|
---|
494 | + " for getColumn is not in required range from 0 up to (but not including) "
|
---|
495 | + numCols + ".");
|
---|
496 | for (var i = 0; i < numRows; i++)
|
---|
497 | A[i, colIndex] = v[i];
|
---|
498 | }
|
---|
499 | /// <summary>
|
---|
500 | /// Sets/Replaces the given row of matrix A with the vector v.
|
---|
501 | /// </summary>
|
---|
502 | /// <param name = "rowIndex">The index of the row, rowIndex.</param>
|
---|
503 | /// <param name = "A">The matrix, A.</param>
|
---|
504 | /// <param name = "v">The vector, v.</param>
|
---|
505 | public static void SetRow(int rowIndex, double[,] A, IList<double> v)
|
---|
506 | {
|
---|
507 | var numRows = A.GetLength(0);
|
---|
508 | var numCols = A.GetLength(1);
|
---|
509 | if ((rowIndex < 0) || (rowIndex >= numRows))
|
---|
510 | throw new Exception("MatrixMath Size Error: An index value of "
|
---|
511 | + rowIndex
|
---|
512 | + " for getRow is not in required range from 0 up to (but not including) "
|
---|
513 | + numRows + ".");
|
---|
514 | for (var i = 0; i < numCols; i++)
|
---|
515 | A[rowIndex, i] = v[i];
|
---|
516 | }
|
---|
517 | /// <summary>
|
---|
518 | /// Makes the zero vector.
|
---|
519 | /// </summary>
|
---|
520 | /// <param name = "p">The p.</param>
|
---|
521 | /// <returns></returns>
|
---|
522 | public static double[] makeZeroVector(int p)
|
---|
523 | {
|
---|
524 | if (p <= 0) throw new Exception("The size, p, must be a positive integer.");
|
---|
525 | return new double[p];
|
---|
526 | }
|
---|
527 |
|
---|
528 | #endregion
|
---|
529 |
|
---|
530 | #region add subtract multiply
|
---|
531 | /// <summary>
|
---|
532 | /// The cross product of two double vectors, A and B, which are of length, 2.
|
---|
533 | /// In actuality, there is no cross-product for 2D. This is shorthand for 2D systems
|
---|
534 | /// that are really simplifications of 3D. The returned scalar is actually the value in
|
---|
535 | /// the third (read: z) direction.
|
---|
536 | /// </summary>
|
---|
537 | /// <param name = "A">1D double Array, A</param>
|
---|
538 | /// <param name = "B">1D double Array, B</param>
|
---|
539 | /// <returns></returns>
|
---|
540 | public static double crossProduct2(IList<double> A, IList<double> B)
|
---|
541 | {
|
---|
542 | if (((A.Count() == 2) && (B.Count() == 2))
|
---|
543 | || ((A.Count() == 3) && (B.Count() == 3) && A[2] == 0.0 && B[2] == 0.0))
|
---|
544 | return A[0] * B[1] - B[0] * A[1];
|
---|
545 | throw new Exception("This cross product \"shortcut\" is only used with 2D vectors to get the single value in the,"
|
---|
546 | + "would be, Z-direction.");
|
---|
547 | }
|
---|
548 |
|
---|
549 | /// <summary>
|
---|
550 | /// The dot product of the two 1D double vectors A and B
|
---|
551 | /// </summary>
|
---|
552 | /// <param name="A">1D double Array, A</param>
|
---|
553 | /// <param name="B">1D double Array, B</param>
|
---|
554 | /// <param name="length">The length of both vectors A and B. This is an optional argument, but if it is already known
|
---|
555 | /// - there is a slight speed advantage to providing it.</param>
|
---|
556 | /// <returns>
|
---|
557 | /// A double value that contains the dot product
|
---|
558 | /// </returns>
|
---|
559 | public static double dotProduct(IList<double> A, IList<double> B, int length)
|
---|
560 | {
|
---|
561 | var c = 0.0;
|
---|
562 | for (var i = 0; i != length; i++)
|
---|
563 | c += A[i] * B[i];
|
---|
564 | return c;
|
---|
565 | }
|
---|
566 | /// <summary>
|
---|
567 | /// The dot product of the one 1D int vector and one 1D double vector
|
---|
568 | /// </summary>
|
---|
569 | /// <param name="A">1D int Array, A</param>
|
---|
570 | /// <param name="B">1D double Array, B</param>
|
---|
571 | /// <param name="length">The length of both vectors A and B. This is an optional argument, but if it is already known
|
---|
572 | /// - there is a slight speed advantage to providing it.</param>
|
---|
573 | /// <returns>
|
---|
574 | /// A double value that contains the dot product
|
---|
575 | /// </returns>
|
---|
576 | public static double dotProduct(IList<int> A, IList<double> B, int length)
|
---|
577 | {
|
---|
578 | var c = 0.0;
|
---|
579 | for (var i = 0; i != length; i++)
|
---|
580 | c += A[i] * B[i];
|
---|
581 | return c;
|
---|
582 | }
|
---|
583 | /// <summary>
|
---|
584 | /// Subtracts one vector (B) from the other (A). C = A - B.
|
---|
585 | /// </summary>
|
---|
586 | /// <param name = "A">The minuend vector, A (1D double)</param>
|
---|
587 | /// <param name = "B">The subtrahend vector, B (1D double)</param>
|
---|
588 | /// <param name="length">The length of the vectors.</param>
|
---|
589 | /// <returns>Returns the difference vector, C (1D double)</returns>
|
---|
590 | public static double[] subtract(IList<double> A, IList<double> B, int length)
|
---|
591 | {
|
---|
592 | var c = new double[length];
|
---|
593 | for (var i = 0; i != length; i++)
|
---|
594 | c[i] = A[i] - B[i];
|
---|
595 | return c;
|
---|
596 | }
|
---|
597 | /// <summary>
|
---|
598 | /// Adds arrays A and B
|
---|
599 | /// </summary>
|
---|
600 | /// <param name = "A">1D double array 1</param>
|
---|
601 | /// <param name = "B">1D double array 2</param>
|
---|
602 | /// <param name="length">The length of the array.</param>
|
---|
603 | /// <returns>1D double array that contains sum of vectros A and B</returns>
|
---|
604 | public static double[] add(IList<double> A, IList<double> B, int length)
|
---|
605 | {
|
---|
606 | var c = new double[length];
|
---|
607 | for (var i = 0; i != length; i++)
|
---|
608 | c[i] = A[i] + B[i];
|
---|
609 | return c;
|
---|
610 | }
|
---|
611 | /// <summary>
|
---|
612 | /// Divides all elements of a 1D double array by the double value.
|
---|
613 | /// </summary>
|
---|
614 | /// <param name="B">The vector to be divided</param>
|
---|
615 | /// <param name="a">The double value to be divided by, the divisor.</param>
|
---|
616 | /// <param name="length">The length of the vector B. This is an optional argument, but if it is already known
|
---|
617 | /// - there is a slight speed advantage to providing it.</param>
|
---|
618 | /// <returns>
|
---|
619 | /// A 1D double array that contains the product
|
---|
620 | /// </returns>
|
---|
621 | public static double[] divide(IList<double> B, double a, int length)
|
---|
622 | { return multiply((1 / a), B, length); }
|
---|
623 | /// <summary>
|
---|
624 | /// Multiplies all elements of a 1D double array with the double value.
|
---|
625 | /// </summary>
|
---|
626 | /// <param name="a">The double value to be multiplied</param>
|
---|
627 | /// <param name="B">The double vector to be multiplied with</param>
|
---|
628 | /// <param name="length">The length of the vector B. This is an optional argument, but if it is already known
|
---|
629 | /// - there is a slight speed advantage to providing it.</param>
|
---|
630 | /// <returns>
|
---|
631 | /// A 1D double array that contains the product
|
---|
632 | /// </returns>
|
---|
633 | public static double[] multiply(double a, IList<double> B, int length)
|
---|
634 | {
|
---|
635 | // scale vector B by the amount of scalar a
|
---|
636 | var c = new double[length];
|
---|
637 | for (var i = 0; i != length; i++)
|
---|
638 | c[i] = a * B[i];
|
---|
639 | return c;
|
---|
640 | }
|
---|
641 |
|
---|
642 | /// <summary>
|
---|
643 | /// Product of a matrix and a vector (2D double and 1D double)
|
---|
644 | /// </summary>
|
---|
645 | /// <param name = "A">2D double Array</param>
|
---|
646 | /// <param name = "B">1D double array - column vector (1 element row)</param>
|
---|
647 | /// <param name="numRows">The number of rows.</param>
|
---|
648 | /// <param name="numCols">The number of columns.</param>
|
---|
649 | /// <returns>A 1D double array that is the product of the two matrices A and B</returns>
|
---|
650 | public static double[] multiply(double[,] A, IList<double> B, int numRows, int numCols)
|
---|
651 | {
|
---|
652 | var C = new double[numRows];
|
---|
653 |
|
---|
654 | for (var i = 0; i != numRows; i++)
|
---|
655 | {
|
---|
656 | C[i] = 0.0;
|
---|
657 | for (var j = 0; j != numCols; j++)
|
---|
658 | C[i] += A[i, j] * B[j];
|
---|
659 | }
|
---|
660 | return C;
|
---|
661 | }
|
---|
662 |
|
---|
663 | public static void multiplyInto(double[,] A, double[] B, int numRows, int numCols, double[] target)
|
---|
664 | {
|
---|
665 | var C = target;
|
---|
666 |
|
---|
667 | for (var i = 0; i != numRows; i++)
|
---|
668 | {
|
---|
669 | C[i] = 0.0;
|
---|
670 | for (var j = 0; j != numCols; j++)
|
---|
671 | C[i] += A[i, j] * B[j];
|
---|
672 | }
|
---|
673 | }
|
---|
674 | /// <summary>
|
---|
675 | /// The cross product of two double vectors, A and B, which are of length, 3.
|
---|
676 | /// This is equivalent to calling crossProduct, but a slight speed advantage
|
---|
677 | /// may exist in skipping directly to this sub-function.
|
---|
678 | /// </summary>
|
---|
679 | /// <param name = "A">1D double Array, A</param>
|
---|
680 | /// <param name = "B">1D double Array, B</param>
|
---|
681 | /// <returns></returns>
|
---|
682 | public static double[] crossProduct3(IList<double> A, IList<double> B)
|
---|
683 | {
|
---|
684 | return new[]
|
---|
685 | {
|
---|
686 | A[1]*B[2] - B[1]*A[2],
|
---|
687 | A[2]*B[0] - B[2]*A[0],
|
---|
688 | A[0]*B[1] - B[0]*A[1]
|
---|
689 | };
|
---|
690 | }
|
---|
691 |
|
---|
692 | #endregion
|
---|
693 |
|
---|
694 |
|
---|
695 | public static double subtractAndDot(double[] n, double[] l, double[] r, int dim)
|
---|
696 | {
|
---|
697 | double acc = 0;
|
---|
698 | for (int i = 0; i < dim; i++)
|
---|
699 | {
|
---|
700 | double t = l[i] - r[i];
|
---|
701 | acc += n[i] * t;
|
---|
702 | }
|
---|
703 |
|
---|
704 | return acc;
|
---|
705 | }
|
---|
706 | }
|
---|
707 | } |
---|